Knowledge-based recommendation uses explicit knowledge about users, items, constraints, and domain rules.
Constraint-Based Recommendation
Constraint-based systems use explicit user requirements.
Examples:
- “Show me laptops under $1000 with 16GB RAM.”
- “Find restaurants within 2 miles that are open now.”
Case-Based Recommendation
Case-based systems recommend items similar to the user’s current need or a previous case.
Use Cases
Knowledge-based recommendation is useful for high-stakes or infrequent decisions:
- cars
- real estate
- computers
- cameras
- complex B2B products
This approach is useful when user preferences change frequently or historical interaction data is weak.